[Opm] Cell Convergence Debugging

OPM User opmuser0 at gmail.com
Mon May 18 08:30:05 UTC 2020


Tony,

Alf is correct, in nearly all cases poor convergence is nearly always 
related to the data: aquifer not in equilibrium with the hydrocarbon 
zone, negative fluid compressibilities, etc. If you can share your model 
then I'll take a look. If not then if you send me the *.PRT, *.LOG and 
*.DBG files I'll have a quick look to see if there is anything obvious 
from these files.

Also are you getting?

|Time step 2, stepsize -nan days, at day 0.04/31, date = 01-Jan-2021 
Error: 
[/build/opm-simulators-4CUPDM/opm-simulators-2019.10-rfinal/opm/simulators/flow/BlackoilModelEbos.hpp:268] 
NaN residual found!�[0m Problem: Solver convergence failure - Numerical 
problem encountered Timestep chopped to -nan days|
||

Nobody  responded to my GitHub issue question "How Many Hours in a -nan 
Day? #2503 " - perhaps it was something of a "black-hole" to investigate 
:-).

I don't' have a general solution to this, but I have resolve a few cases 
of the above due to data inconsistencies.

Release 2020-04 is out, so perhaps you can try your model with the new 
release.
||

OPM User
opmuser0 at gmail.com




More information about the Opm mailing list